Integrative Proteomics–Metabolomics Strategy for Pathological

Nov 30, 2017 - School of Life Sciences and Technology, Shanghai Tech University, Shanghai 200433, China. ⊥ Changhai Hospital, Second .... After the ...
2 downloads 12 Views 4MB Size
Subscriber access provided by READING UNIV

Article

Integrative Proteomics-Metabolomics Strategy for Pathological Mechanism of Vascular Depression Mouse Model Hongxia Zhao, Hongli Du, Min Liu, Songyan Gao, Na Li, Yufan Chao, Ruiqing Li, Wei Chen, Ziyang Lou, and Xin Dong J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.7b00724 • Publication Date (Web): 30 Nov 2017 Downloaded from http://pubs.acs.org on December 2, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

graphic abstract 170x177mm (300 x 300 DPI)

ACS Paragon Plus Environment

Journal of Proteome Research

Page 2 of 37

Integrative Proteomics-Metabolomics Strategy for Pathological Mechanism of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Vascular Depression Mouse Model Hongxia Zhaoa1, Hongli Dub1, Min Liuc1, Songyan Gaoa, Na Lia, Yufan Chaoa, Ruiqing Lid, Wei Chen e*, Ziyang Loua*, Xin Donga**.

a

School of Pharmacy, Second Military Medical University, Shanghai 200433, China.

b

department of Pharmacy, Eastern Hepatobiliary Surgery Hospital, Shanghai200433, China.

c

Pharmacy departent of Changhai Hospital, Second Military Medical University, shanghai200433, China.

d

e

1

School of Life Sciences and Technology, ShanghaiTech University, Shanghai, China. Changhai Hospital, Second Military Medical University, Shanghai 200433, China

These authors contributed equally to this work.

Corresponding author: e*E-mail: [email protected]; a*E-mail:[email protected]; a** E-mail: [email protected], Fax: +86 21- 81871335; Tel: +86 21-81871335

Key words: Proteomics; Metabolomics; Vascular depression; LC-MS Abbreviations VD

Vascular depression

GCI

Global cerebral ischemia

PSD

Post-stroke depression

iTRAQ

Isobaric Tags for Relative and Absolute Quantitation

SPF

Specific pathogen Free

OFT

Open-field test

TST

Tail suspension test

FST

Forced swimming test Ultra-performance liquid chromatography-quadruple-time

UPLC-Q-TOF-MS flight tandem mass spectrometry

1 / 36

ACS Paragon Plus Environment

of

Page 3 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

DCX

Doublecortin

GFAP

Glial fibrillary acidic protein

PCA

Principal component analysis

PLS-DA

Partial least squares-discriminate analysis

IPA

Ingenuity pathway analysis

OEA

N-oleoyl ethanolamine

ACN

Acetonitrile

BCA

Bicinchonininc acid

SDS-PAGE

SDS-polyacrylamide gelelectrophoresis

MCAO

Middle cerebral artery occlusion

TIC

Total ion chromatography

EIC

Extracted ion chromatography

GAPDH

Glyceraldehyde-3-phosphate

NMDA

N-methyl-D-aspartic acid receptor

AMPA

α-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid

MCI

Mild

GO

Gene ontology

VIP

Variable importance

MRM

Multiple reaction monitor

LPT

Long Term Potentiation

NAA

N - acetyl - L-aspartic acid

NAAG

N-acetyl-L-aspartic acid-L-glutamic acid

NAALADase

N-acetylated alpha-linked-acidic enzyme

CNS

Central nervous system

ROS

Reactive oxygen species

cognitive impairment

2 / 36

ACS Paragon Plus Environment

Journal of Proteome Research

Page 4 of 37

Abstract 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Vascular depression(VD), a subtype of depression, is caused by vascular diseases or cerebrovascular risk factors. Recently, the proportion of VD patients has increased significantly, which severely affects their quality of life. However, the current pathogenesis of VD has not yet been fully understood, and the basic research is not adequate. In this study, based on the combination of LC-MS-based proteomics and metabolomics, we aimed to establish a protein metabolism regulatory network in a murine VD model to elucidate a more comprehensive impact of VD on organisms. We detected 44 metabolites and 304 proteins with different levels in the hippocampus samples from VD mice using a combination of metabolomic and proteomics analyses with an isobaric tags for relative and absolute quantification (iTRAQ) method. We constructed a protein-to-metabolic regulatory network by correlating and integrating the differential metabolites and proteins using ingenuity pathway analysis. Then we quantitatively validated the levels of the bimolecules shown in the bioinformatics analysis using LC-MS/MS and Western blotting. Validation results suggested changes in the regulation of neuroplasticity, transport of neurotransmitters, neuronal cell proliferation and apoptosis, and disorders of amino acids, lipids and energy metabolism. These proteins and metabolites involved in these dis-regulated pathways will provide a more targeted and credible direction to study the mechanism of VD. Therefore, this paper presents an approach and strategy that was applied in integrative proteomics and metabolomics for research and screening potential targets and biomarkers of VD, which could be more precise and credible in a field lacking adequate basic research.

Introduction Depression is a common mental health problem in modern society, and the symptoms are mainly manifested as sadness and loss of interest or fun in surrounding activities. Patients often experience guilt, low self-esteem, low sleep quality, loss of appetite, fatigue, and reduction in attention, and sometimes commit suicide 1. Vascular depression (VD) is a subtype of depression, first proposed by Alexopoulos et al. 2. VD is defined as depression caused by vascular disease or cerebrovascular risk factors 3 / 36

ACS Paragon Plus Environment

3-7

. Statistics have shown that

Page 5 of 37

Journal of Proteome Research

up to 33% of cerebral ischemia patients caused by myocardial infarction will develop depression 8. The 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

proportion of patients with VD in the total population of patients with depression is increasing, and studies have found that patients with VD accounted for 1/5 of adult depression patients 9. In addition, VD accounts for 50% of patients with severe depression 10. Recent clinical studies have shown that the elderly are more prone to VD, resulting in not only an increase in cognitive dysfunction, but also a serious negative impact on the prognosis 11, 12

. At the same time, VD will increase the risk of cerebrovascular abnormalities and even stroke 13. Population

aging, a fast pace of life, and increasing stress have resulted in a high occurrence of cerebrovascular diseases, which also leads to an increased incidence of VD; meanwhile, VD is usually interrelated with high blood pressure disease; therefore, the treatment strategy for VD is more complicated than that of other types of depression 14. Currently, the vast majority of studies on depression primarily focus on severe depression, and the diagnosis of VD and the corresponding treatment strategy are undefined

15

because of difficulties in the

clarification of the relevant mechanisms. Therefore, it is imperative to conduct studies with greater precision and credibility that would benefit future studies on VD, which is a clinically common disease. Systems biology, which elucidates the physiological state of an organism by integrating biological information at different levels, such as various interactions between genetic elements and proteins, and metabolic and regulatory pathways2, is very suitable for establishing early directional contours for a follow-up study. In the post-genome era, system biology focuses on "functional interpretation", and proteomics and metabolomics are its main research approaches. To have a comprehensive and in-depth understanding of complex biological activities of life, proteomic studies are very useful. With the development of liquid chromatography-mass spectrometry technology, modern proteomics techniques, such as quantitative proteomics using iTRAQ markers, have greatly improved the detection abilities and reproducibility

16

, and it

has been widely used in a variety of studies on the molecular markers and mechanisms of diseases

17, 18

.

Additionally, the metabolites produced during life activities directly and accurately reflect the pathophysiology of organisms. Metabolomics studies all the endogenous metabolites in organisms or cells and reveals their changes and is considered to be an extension and supplementation of genomics and proteomics. The

4 / 36

ACS Paragon Plus Environment

Journal of Proteome Research 19-21

combination of proteomics and metabolomics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 37

can provide direct evidence and a clearer explanation for the

changes in organisms. Therefore, this study used bilateral ligation of the common carotid artery to induce global cerebral ischemia (GCI) to build a model of VD in mice

22, 23

. Then, by employing proteomics and metabolomics

analyses based on LC-MS, we conducted integrative studies on mouse hippocampus tissue to establish an overall regulatory protein-metabolite network related to VD, and we validated the levels of the molecules involved in the pathways revealed by our bioinformatics study. This whole experimental flow chart is shown in Fig. 1. The results suggest a number of biological pathways associated with VD, which provide a more targeted and more reliable research direction for future study of the mechanisms of VD. Moreover, these biomolecules could be potential therapeutic targets or biomarkers for VD. [Fig. 1] Experimental procedures Chemicals and reagents-Acetonitrile (LC/MS grade) and methanol (HPLC grade) were purchased from Merck (Darmstadt, Germany). Formic acid, niacinamide, L-aspartic acid, hypoxanthine, L-methionine, xanthine, L-phenylalanine, L-arginine, tryptophan, inosine, oleoylethanolamide and other standard chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA). Antibodies were purchased from Abcam Corporation (Abcam, UK). An iTRAQ kit was purchased from AB SCIEX (AB SCIEX, Foster City, CA, USA) and other reagents (analytic quality) were commercially available. Animal model, behavior studies, and sample collection Global cerebral ischemic (GCI) Depression Model Building- A total of 57 SPF grade ICR mice (18-20g) were purchased from Shanghai Super-B&K Laboratory Animal Corp. Ltd. and maintained at 20-23℃ and a humidity of 40-60%, with a 12h light/dark cycle. Animals had free access to food and water. After one week of adaption to the environment, all mice were randomly divided into two groups: 26 in the sham group and 31 in the GCI model group. We established a model of VD based on a previous model of ischemia-reperfusion injury 23

. The sham group was treated the same way except the blood supply was blocked. After the operation, the

5 / 36

ACS Paragon Plus Environment

Page 7 of 37

Journal of Proteome Research

mouse’s normal body temperature was maintained using an electric heater until the mice could adjust their body 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

temperature. The mice were kept for 3 days to heal the wounds before participating in the behavior tests. The numbers of surviving animals in both the final sham group and the GCI group were 26. Behavior tests- The weight of each mouse was recorded at the same time every two days. An open-field test (OFT) 24, tail suspension test (TST) 25, and forced swimming test (FST)

26

were conducted as described in the

previous literature. The results of OFT were evaluated by the total number of horizontal passes and the number of vertical standing times. The TST and FST were examined by the time the animals became immobile. Then, we used SPSS software to analyze the results of the behavioral experiments using independent sample T tests. Sample collection and immunofluorescence-All animals were given anesthesia and immediately decapitated after the behavioral experiments. The hippocampus was dissected out on ice. Among the whole sample of each group, the hippocampus of 6 mice was used for proteomics analysis and Western blot analysis, then 6 were used for metabolomics analysis and 10 for metabolites targeted validation. These samples were quickly frozen in liquid nitrogen and stored at -80 ℃ for further study. The last 4 samples were fixed in 4% paraformaldehyde overnight for immunofluorescence and then incubated with 30% glucose solution for cryopreservation. The consecutive frozen sections (20 µm) of the hippocampus along the coronal plane were incubated with goat anti-DCX antibodies (Santa Cruz,1: 500) and rabbit polyclonal GFAP (Abcam,1: 1000) overnight at 4 ℃. After washing with PBS, the sections were incubated with the corresponding secondary antibody Alexa Fluor594 donkey anti-goat IgG (Abcam,1 : 500) and Alexa Fluor 488 goat anti-rabbit IgG (Abcam,1: 500) and then incubated at 37 ℃ for 2 h in the dark. The images were observed and acquired under immunofluorescence microscopy. Metabolomics analysis- The metabolomics study was based on previously reported methods27. The specific experimental procedure is attached in a Supplementary file A. To verify the accuracy of the identities of metabolites whose levels were different between the two groups, we first confirmed the exact molecular weights of these metabolites using an extracted ion flow chromatograph (EIC) and then compared the exact molecular weights with the common online databases, such as the Human Metabolome Database 6 / 36

ACS Paragon Plus Environment

Journal of Proteome Research

Page 8 of 37

(http://www.hmdb.ca/) and Metlin (http://metlin.scripps.edu). We also compared the retention times of some 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

metabolites and their fragments in MS/MS with their standards and fragments under the same conditions and compared MS/MS fragments of other metabolite molecules with the MS/MS fragments in the database above to correctly identify these metabolites. The relative levels of metabolites between the groups are shown in a heat map analysis based on the MetaboAnalyst platform (http://www.metaboanalyst.ca). Proteomic Analysis- The quantitative analysis of iTRAQ proteomics was based on a previous report 28. Half of each hippocampus of the mice in each group was randomly taken and every 3 half-hippocampus were randomly pooled into one sample in each group. Each group undergo the same pre-process and labled with 113,114 tags for sham group and 115,116 tag for model group. Samples were analyzed by the eksigent nano LC-UltraTM system tandem TripleTOFTM5600 mass spectrometer (AB SCIEX, Foster City, CA, USA). Using an auto sampler, 4μl of sample was loaded onto a Nano LC trap column (ChromXP C18_CL-3μm,120A,350μm *0.5mm) and eluted with solvent A (2% acetonitrile with 0. 1% (v/v) formic acid) at 3μL/min for 15 min and separated on the analyzed colume(0.075×150mm,3µm,120A) with a linear gradient of solvent A and solvent B (98% acetonitrile with 0. 1% (v/v) formic acid) at a flow rate of 300 nL/min over 120 min: 0-0.1 min,5-10% B; 0.1-80 min,10-23% B;80-105 min,23-35% B;105-106 min,35-80% B;106-110 min,80% B;110-110.5 min,80-5% B;110.5-120 min,5% B. Mass spectrometry condition isas follow: using ESI ion source (SilicaTipTM FS360-20-10-N-20-C12,Tip:10±1μm,USA.) in positive condition; and ion data in m/z range of 350-1250Da was selectively acquired. Curtain gas: 30psi; atomized gas: 15psi; ion atomization voltage: 2.3kv; ESI temperature: 150℃; accumulative scan time in high resolution scan mode is 250ms;

MS/MS

spectra of at most 40 precursor are acquired in every cycle time and the charge of fragmentors are +2~+5. iTRAQ adjust rolling collision energy was used during the whole process. Tandem mass spectrometry data were analyzed using an AB Sciex MS data converter (version 1.3) followed by using Mascot (Matrix Science, London, UK; version 2.5.1).

The mouse total protein database was downloaded from Uniprot

(http://www.uniprot.org/). Peptides and proteins were identified using Scaffold (version Scaffold_4.6.1, Proteome Software Inc., Portland, OR). At the FDR 99%. The quantitative analysis of peptides and proteins was 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

primarily carried out using Scaffold Q + (version Scaffold_4.6.1, Proteome Software Inc., Portland, OR) software. First, the sample peaks were iteratively normalized. In this experiment, we used the Mann-Whitney Test to compare the differences between two groups. after the correction for multiple hypothesis test, we found that proteins whose P value less than 0.00833 were much more credible. Finally, we determined the numbers of the proteins whose statistically significant fold changes between groups were above ±1.3 29. These proteins with differential

levels

between

the

groups

were

then

analyzed

by

Gene

Ontology

(GO)

(http://www.geneontology.org/) for differential protein enrichment at the levels of both the “biological process” and “cellular component”. Bioinformatics Analysis of Metabolites and Proteins- To investigate the mechanism of VD, we performed metabolomics profile analysis and proteomics analysis of differential metabolites by using QIAGEN's Ingenuity Pathway Analysis (IPA®, QIAGEN Redwood City, www.qiagen.com/ingenuity) which evaluate the results mainly by two parameters(p value and z-score), p value is based on Right-Tailed Fisher’s Exact Test algorithm, it shows whether the association between a set of meaningful molecules in your experiment and the known process / pathway / transcription comes from random matching , it neither takes into account the effects of molecules nor fold change between molecules in the data set. While z-score evaluate the effect of molecular changes on biological processes, in general, z-score>2 suggests that the corresponding molecules / functions are significantly activated, Z< -2 thinks that the corresponding molecules / functions are significantly inhibited. Validations of the differently expressed proteins and metabolites - To demonstrate the reliability of the differences in the levels of metabolites between the groups obtained by mass spectrometry, we quantitatively validated 10 metabolites. The specific experimental procedure is attached in the Supplementary file B. Table 1 shows the optimized MRM parameters for each target metabolite. The standard curves were generated using the standards diluted in 80% ACN. In addition, 100ng of L-chlorophenylalanine was added as an internal standard in other standards except(N-oleoylethanolamine, OEA) and samples. Student’s t test was used to analyze the significant differences between the two groups. Among them, the levels of all metabolites were normalized with the internal controls and each sample mass. 8 / 36

ACS Paragon Plus Environment

Journal of Proteome Research

Page 10 of 37

Moreover, to verify the accuracy of the results obtained by mass spectrometry in proteomics analysis, we 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

examined some proteins with different levels in the two groups by Western blotting. First, the protein samples(n=3, the rest parts of every two mice was mixed into one sample, six mice randomly divided into three samples.) were prepared the same way as for the proteomics experiment described above. Nine proteins were identified by Western blotting, namely, Camk2a, Ppp1r1a, Prkcb, Gria1, Grin2b, Gria2, Prkcg, Gnaq, and Slc17a7, and the corresponding antibodies (ab92332, ab40877, ab195039, ab109450, ab183942, ab133477, ab108961, ab199533, and ab180188) were from Abcam (Abcam, UK). The experimental procedures were as reported previously

30

. The signals were visualized with Ondyssey® Imager (LI-COR, Lincoln, NE, USA),

gray scanning analysis were carried out after normalized by internal reference GAPDH and use independent sample T-test to analyze the difference with two groups. Results Behavioral analysis- Deceleration of activity, loss of appetite and weight loss seem to be the manifestations of somatic symptoms of depression in mice. The motions of mice slow down. The mouse body weight change curve is shown in Fig. 2A. Within one week of adaptive feeding, there was no difference in body weight between the two groups. However, after the ischemic procedures, the difference was significantly increased with prolongation. At the time point of tissue collection, the body weights of the experimental group were significantly lower than those of the sham group (p